Main theme of the patch is to fix joins processing in the plugin
code. We now use SELECT_LEX::top_join_list and process the nested
joins recursively, instead of SELECT_LEX::table_list struct which
we earlier used to build the join filters. The earlier approach
did not process certain nested join ON expressions, causing certain
queries to incorrectly error out such as that described in MCOL-4680.
In addition, some legacy code is also removed.
We earlier leveraged the server functionality provided by
Item_in_subselect::create_in_to_exists_cond and
Item_in_subselect::inject_in_to_exists_cond
to create and inject the in-to-exists predicate into an IN
subquery's JOIN struct. With this patch, we leave the IN subquery's
JOIN unaltered and instead directly perform this predicate creation
and injection into ColumnStore's select execution plan.
For a query of the form:
SELECT COUNT(c2) FROM (SELECT * FROM t1) q;
where t1 contains 10 columns c1, c2, ... , c10.
We currently create an intermediate RowGroup in ExeMgr with
a row of the form (1, c2_value1, 1, 1, 1, 1, 1, 1, 1, 1), i.e.
for all the columns of the subquery which are not referenced in
the outer query, we substitute a constant value, which is wasteful.
With this optimization, we are trimming the RowGroup to a row
of the form (1, c2_value1). This can have non-trivial query
execution time improvements if the subquery contains large number
of columns (such as a "select *" on a very wide table) and the outer
query is only referencing a subset of these columns with lower
index values from the subquery (as an example, c1 or c2 above).
That is, the current limitation of this optimization is we are not
removing those non-referenced subquery columns (c1 in the query above)
which are to the left of a referenced column.
After the cleanup work done for FETCH FIRST ... WITH TIES
SELECT_LEX members select_limit, explicit_limit and offset_limit are now moved
to SELECT_LEX::limit_params.
type is not LEFT/RIGHT.
In buildOuterJoin(), do not add ON expressions for WHERE
processing when the JOIN type is not LEFT/RIGHT.
Test cases to check correct processing of INNER JOIN ON expressions
with possible/impossible WHERE conditions are added for
1. One side of the LEFT JOIN being INNER JOIN.
2. One side of the LEFT JOIN being an INNER JOIN inside an INNER JOIN.
3. Both sides of the LEFT JOIN being an INNER JOIN.
For TIMESTAMP, it should do similar. However, it didn't work. For some reason, MDB has the function set as DATETIME, which for cs, isn't the same thing. Added a kludge to ha_mcs_execplan.cpp to handle it.
1. Add wide decimal support to AggregateColumn::evaluate
and TreeNode::getDecimalVal().
2. Use the pm aggregate attributes to determine um aggregate
attributes in TupleAggregateStep::prep2PhasesAggregate.
1. Perform type promotion to wide decimal if the result
of an arithmetic operation has a precision > 18.
2. Only set the decimal width of an arithmetic operation to wide
if both the LHS and RHS of the operation are decimal types.
Introduced fDecimalOverflowCheck to enable/disable overflow check.
Add support into a FunctionColumn.
Low level scanning crashes on medium sized data sets.
2. Set Decimal precision in SimpleColumn::evaluate().
3. Add support for int128_t in ConstantColumn.
4. Set IDB_Decimal::s128Value in buildDecimalColumn().
5. Use width 16 as first if predicate for branching based on decimal width.
TupleAggregateStep class method and buildAggregateColumn() now properly set result data type.
doSum() now handles DECIMAL(38) in approprate manner.
Low-level null related methods for new binary-based datatypes now handles magic values for
binary-based DT.
Create tables and schemas with lower case name only if the flag is set.
During operations, convert to lowercase in plugin. Byt the time a query gets to ExeMgr, DDLProc etc., everything must be lower case if the flag is set, and undisturbed if not.
This patch enables select handler for executing prepared
statements. Most importantly, we are now activating a
persistent arena which will allocate any new items in a
permanent MEMROOT for prepared statements and stored procedures.
Refer to JOIN::optimize_inner() for details.
In processWhere(), we now use SELECT_LEX::prep_where in case
we are executing a prepared statement, as this is where the saved
WHERE clause is stored for prepared statement processing.
In addition, we also disable derived handler for prepared
statements.
For certain queries, such as:
update cs1 set i = 41 where i = 42 or (i is null and 42 is null);
the SELECT_LEX.where does not contain the required where conditions.
Server sends the where conditions in the call to cond_push(), so
we are storing them in a handler data member, condStack, and later
push them down to getSelectPlan() for UPDATES/DELETEs.
